Sample
Letters: Initial Appointment
Sample
Dean's Recommendation - Initial Appointment
April 15, 1996
PROVOST EDWARDS:
Chair Smith of the Statistics Department proposes the appointment of Jane Roe as Assistant Professor, Step II, effective July 1, 1996. Roe finished her Ph.D. in theoretical statistics in January, 1996 at the University of Manchester under the supervision of William Rogers and is currently engaged in the applied mathematics research program at Michigan. Reviewer A ranks her "as one of the best probabilists 'just out," and the departmental committee is strongly impressed by Roe's accomplishments so far. These accomplishments include six research papers (two published, two accepted, two submitted) and a variety of related activities in the applied statistics area; she also shows considerable promise as a teacher.
Roe has accepted a one-year appointment in Michigan. The Statistics Department is confident that she will come to Berkeley in 1997 or before if we make the proposed appointment. Roe is stronger in promise than other recent appointments in the Division and appears to justify the unusual effort required to secure her services. I support Roe's appointment as Assistant Professor, Step II, effective July 1, 1996, on leave without salary for the 1996-97 academic year.
April 3, 1996
Dear Dean Jones,
Re: Appointment of Jane Roe as Assistant Professor
This letter is a request that Jane Roe be appointed as Assistant Professor, Step II, in the Department of Statistics, effective July 1, 1996. Roe is the candidate chosen by Statistics after a very extensive tenure-track search this academic year. Her appointment was discussed at a department meeting on March 30, 1996. The results of the subsequent secret ballot of the regular faculty were: 15 in favor, two against and four absent. The two negative votes were cast by faculty members who preferred other candidates. The four absent members were abroad and could not be reached.
Roe is the stronger of two excellent finalists for our advertised tenure-track position. The other finalist has been appointed as a Visiting Assistant Professor here for 1996-97. Roe received her Ph.D. in January from the University of Manchester under the supervision of William Rogers, one of the foremost theoretical statisticians in the world. Roe has completed and submitted for publication six research papers. Two of these papers have already been published in distinguished refereed journals, and two have been accepted for publication in another such journal.
The content and remarkable quality of Roe's papers are discussed in the accompanying review by Reviewers B and C. They conclude: "Roe has done very impressive work and managed to get it published in a very short time . . . She is an expert in several different branches of modern probability theory. She has made progress gaining international recognition in an area already well worked, and she has opened up the completely new area of local distance measures in nonparametric settings." Roe's extraordinarily substantial early research achievement is the basis for the proposed appointment.
Three outside letters from leading statisticians and probabilists with international reputations firmly support Roe's candidacy. Reviewer A writes: "From the beginning, it was clear that she is very hard working and able. She reads widely and generates her own ideas. Almost all the topics she worked on during her Ph.D. arose from her own reading and thinking...Roe must be ranked as one of the best statisticians 'just out', and I am sure she will produce some exciting work in the next few years." Reviewer D writes: "I find the thesis very impressive...it exploits such an array of powerful techniques so fully as to obtain substantial results, some of which (as in the case of the Briller conjecture) have defeated masters of the subject....." Reviewer E says that he would "expect Roe to be a good teacher for she has a clear mind and relates well to people." They add that Roe's research on estimation theory based on local distance measures is a remarkably sustained investigation for a research student to have undertaken, and she has achieved extraordinarily sharp results."
The fine seminar that Roe gave here in late February confirmed her ability as an expositor of advanced technical ideas. I fully expect her teaching will be very successful. She has unusually broad interests which range form biomedical statistics to probability theory.
Roe is a first-rate candidate for a tenure-track Assistant Professorship. She has already accepted a one year Johnson Instructorship at the University of Michigan with Tolerill's strong group. In addition, she was offered a tenure-track position at Cambridge and a visiting position at Stanford. Discussions with her indicate that she would very much like to join this department and would be willing to make a commitment now for July 1, 1997. I therefore urge the College to offer Roe a position as Assistant Professor, Step II, effective July 1, 1996, with the understanding that she could take up to a year's leave-of absence without pay before coming to Berkeley.
March 18, 1996
TO: Charles Smith, Chair
FROM: Reviewers B and C (Search Committee)
RE: The case for hiring Jane Roe for tenure track position in theoretical statistics
This is very strong case. Roe started her career with rigorous undergraduate training in mathematics and statistics at Cambridge University. There she had an extraordinary record of straight high distinctions leading to first class honors. Roe then went to Manchester where she was supervised by William Rogers, one of England's foremost theoretical statisticians. Roe's research work at Manchester went deep into the mathematics of differential geometry with application to estimation theory, work for which she has already acquired international recognition. Here is a review of her research performance, awards and reputation, followed by an account of her teaching and a summary of the case.
Research:
Roe's bibliography shows two papers (1 and 2) published by refereed journals with international reputations, as well as two papers (3 and 4) accepted, and two further papers (5 and 6) submitted to journals of similar distinction. All the papers were derived from her very substantial Manchester Ph.D. thesis. Paper 7, which is in draft form and has not as yet been submitted for publication, is subsequent to her doctoral work.
Paper 1 on the Lendel dimension of cone points, concerns the sets of times at which a Brownian path in two or more dimensions stays inside a cone. This paper gives much more precise information than the results of Takahashi and Fenton, and is an incisive piece of work.
Paper 2 is on approximation theory with application to second order efficient estimates and is written jointly with her adviser at Cambridge. This paper unifies a body of disparate results on the convergence of approximation procedures and establishes asymptotic results under weaker moment conditions than previously known.
Paper 3 develops local distance measures in a parameter free setting. Not only was Roe able to improve on earlier results, she did so by placing them in the context of a general theory.
Paper 4 on Hendrick's family of estimates is a further development of paper 3. Roe also obtains a result similar to Lecault and Renwick but Roe's result is slightly stronger and her proof more direct. These three papers (2-4) display a mastery of the field of efficient estimates and opens up the whole area in an unexpected and exciting fashion. The mathematics used is extraordinarily elegant for someone at the beginning of her career.
Papers 5, 6, and the draft of 7 concern sample properties of Maxwell processes indexed by local fields. This work takes off from the well established theory of Maxwell processes indexed by a metric space and shows how much sharper results can be obtained in the local field by exploiting decompositions into disconnected components. Theorists and probabilists in the department, particularly Reviewers P & Q and ourselves, were very impressed with the novelty of the results and the depth and thoroughness of her analysis. These papers illustrate her ability to develop independent lines of inquiry and formulate her own research agenda. It is also important to note that her most recent work was motivated by discussions with her applied colleagues at Michigan suggesting that she is well placed to continue interdisciplinary work here at Berkeley.
Awards and Reputation:
As an undergraduate at Cambridge, Roe won a succession of prizes for the top student in her year, most outstanding student in her college and finally in 1988 the University Medal in Mathematics.
Recently Roe accepted a Johnson Research Instructorship at the University of Michigan. This is a distinguished position with reduced teaching load, which she obtained in competition with candidates in all branches of mathematics. This is a very attractive position for her as she will be able to work with Martin Tollerill, the foremost theoretical statistician in the world.
Teaching:
In her final year in Cambridge, Roe was employed by the Department of Mathematics and Statistics as part-time tutor, teaching a service course to a class of about 40 undergraduate chemists. Reviewer R reports that Roe was a very conscientious teacher. We have no formal classroom surveys but her colleagues at Cambridge were quick to praise her organization, attention to detail and her delivery.
All of her referees say that she will make a good lecturer, and we have the same impression based on her seminar last week. Her account of her research was particularly clear and well organized. By training she is competent to teach the subjects of just about all our undergraduate offerings. And we think she could cope with even our biggest lower division courses.
Summary:
Roe has done very impressive work and managed to get it published in a very short time. She was only three years at Manchester. She is an expert in several different branches of modern probability theory. She has made progress gaining international recognition in areas already well worked, and she has opened up the completely new area of local distance measures in nonparametric settings. She is completely in command of an arsenal of powerful mathematical tools that are good for many problems other than those to which she is currently applying them: e.g., Fourier spectral analysis of timed sequences, potential theory and inverse problems, Strassberg measures, reproducing Kemel Hilbert spaces. She has done first rate work already and she looks set to do lots more.
This appointment, we believe, is an unusually promising one for the department. Here is someone very smart, with a solid background in mathematics and statistics, a strong research record, versatile teaching competence, a breadth of research interests and application, and capability of doing research on just about any problem that interests her.
Chair's & Dean's appointment letters:
|
Reviewer Code |
Reviewer Name |
Affiliation |
|
Reviewer A |
Professor W. Rogers |
University of Manchester |
|
Reviewer B |
Professor L. Johns |
UCB |
|
Reviewer C |
Professor T. Smythe |
UCB |
|
Reviewer D |
Professor S. Weber |
Cambridge University |
|
Reviewer E |
Professor M. Tolerill |
University of Michigan |
Search Committee Report:
|
Reviewer Code |
Reviewer Name |
Affiliation |
|
Reviewer P |
Professor J. Mills |
UCB |
|
Reviewer Q |
Professor A. Kersey |
UCB |
|
Reviewer R |
M. Thomas, Sr. Lecturer |
Cambridge University |
Page last checked: 10/20/99